High Correct Recognition Rate Classifier Design with Appropriate Rejection Mechanism

YANG Guo-wei; WAN Ming-hua; LAI Zhi-hui; ZHANG Fan-long; WONG Wai-keung

doi:10.12263/DZXB.20190518

您当前的位置：

首页 >

文章列表页 >

High Correct Recognition Rate Classifier Design with Appropriate Rejection Mechanism

PAPERS | 更新时间：2025-12-08

- High Correct Recognition Rate Classifier Design with Appropriate Rejection Mechanism
- ACTA ELECTRONICA SINICA Vol. 49, Issue 8, Pages: 1569-1576(2021)
- 作者机构：
  
  1.青岛大学电子信息学院，山东青岛 266071
  2.南京审计大学信息工程学院，江苏南京 211815
  3.深圳大学计算机与软件学院，广东深圳 518060
  4.香港理工大学，香港九龙 999077
- 作者简介：
- 基金信息：
- DOI：10.12263/DZXB.20190518
  CLC： TP3;
- Received：13 May 2019，
  
  Revised：2020-11-19，
  
  Published：25 August 2021
- 稿件说明：
移动端阅览
杨国为,万鸣华,赖志辉等.具有合适拒识机制的高正确识别率分类器设计[J].电子学报,2021,49(08):1569-1576.

YANG Guo-wei,WAN Ming-hua,LAI Zhi-hui,et al.High Correct Recognition Rate Classifier Design with Appropriate Rejection Mechanism[J].ACTA ELECTRONICA SINICA,2021,49(08):1569-1576.
杨国为,万鸣华,赖志辉等.具有合适拒识机制的高正确识别率分类器设计[J].电子学报,2021,49(08):1569-1576. DOI： 10.12263/DZXB.20190518.

YANG Guo-wei,WAN Ming-hua,LAI Zhi-hui,et al.High Correct Recognition Rate Classifier Design with Appropriate Rejection Mechanism[J].ACTA ELECTRONICA SINICA,2021,49(08):1569-1576. DOI： 10.12263/DZXB.20190518.

摘要

针对目前一些正确识别率高的SVM（Support Vector Machines）分类器、超球SVM分类器、深度学习分类器在一些典型样本集上应用时仍然有2%左右的错误识别率和增量学习功能不强的问题，本文提出了一种具有合适拒识机制的高正确识别率分类器设计方案和相应的增量学习算法，较好地解决了上述问题.主要工作包括：同类特征集合的紧密包裹集构造算法；基于同类特征集合和紧密包裹集的同类特征区域紧密包裹面的求解算法；设置所有紧密包裹面之外的公共区域为分类器的拒识区域的方法；当增加新类别、增减训练样本时，以上算法的增量学习算法.用uci数据集做对比实验表明，在拒识率小于1.3%的情况下，本文方法设计的分类器正确识别率大于99.13%.

Abstract

At present

some SVM(Support Vector Machines) classifiers

hypersphere SVM classifier and deep learning classifier with high correct recognition rate still have about 2% false recognition rate and weak incremental learning function. In this paper

a high correct recognition rate classifier designed with appropriate rejection mechanism and incremental learning algorithm is proposed to solve the above problems. The main work include: the construction algorithm of compact packing set of homogeneous feature set; the algorithm for solving the compact packing surface of homogeneous feature region based on homogeneous feature set and compact packing set; the method of setting all the public areas outside the compact packing surface as the rejection area of the classifier; when adding new categories

increasing or decreasing training samples

the above algorithms are incremental learning algorithms. A comparison experiment with uci data sets shows that the correct recognition rate of the classifier is greater than 99.13%

when the rejection rate is less than 1.3%.

关键词

Keywords

references

王守觉 . 仿生模式识别(拓扑模式识别): 一种模式识别新模型的理论与应用 [J]. 电子学报 , 2002 , 30 ( 10 ): 1417 － 1420 .

Wang S J . Bionic (topological) pattern recognition—A new model of pattern recognition theory and its applications [J]. Acta Electronica Sinica , 2002 , 30 ( 10 ): 1417 － 1420 . (in Chinese)

周志华 . 机器学习 [M]. 北京 : 清华大学出版社 , 2016 .

Zhou Z H . Machine Learning [M]. Beijing, China : Tsinghua University Press , 2016 . (in Chinese)

Vapnik V N . The Nature of Statistical Learning Theory [M]. New York, NY , USA: Springer New York , 1995 .

Tang F Z , Adam L , Si B L . Group feature selection with multiclass support vector machine [J]. Neurocomputing , 2018 , 317 ( 23 ): 42 － 49 .

Tian Y J , Qi Z Q , Ju X C , et al . Nonparallel support vector machines for pattern classification [J]. IEEE Transactions on Cybernetics , 2014 , 44 ( 7 ): 1067 － 1079 .

Ding L Z , Liao S Z . An approximate approach to automatic kernel selection [J]. IEEE Transactions on Cybernetics , 2017 , 47 ( 3 ): 554 － 565 .

Cano A , Zafra A , Ventura S . Weighted data gravitation classification for standard and imbalanced data [J]. IEEE Transactions on Cybernetics , 2013 , 43 ( 6 ): 1672 － 1687 .

Jayadeva , Khemchandani R , Chandra S . Twin support vector machines for pattern classification [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2007 , 29 ( 5 ): 905 － 910 .

Xu Y T , Yang Z J , Pan X L . A novel twin support-vector machine with pinball loss [J]. IEEE Transactions on Neural Networks and Learning Systems , 2017 , 28 ( 2 ): 359 － 370 .

Wu M R , Ye J P . A small sphere and large margin approach for novelty detection using training data with outliers [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2009 , 31 ( 11 ): 2088 － 2092 .

Xu Y T , Liu C M . A rough margin-based one class support vector machine [J]. Neural Computing and Applications , 2013 , 22 ( 6 ): 1077 － 1084 .

Chorowski J , Wang J , Zurada J M . Review and performance comparison of SVM- and ELM-based classifiers [J]. Neurocomputing , 2014 , 128 : 507 － 516 .

Xu Y T , Yang Z J , Zhang Y Q , et al . A maximum margin and minimum volume hyper-spheres machine with pinball loss for imbalanced data classification [J]. Knowledge-Based Systems , 2016 , 95 : 75 － 85 .

Xu Y T . Maximum margin of twin spheres support vector machine for imbalanced data classification [J]. IEEE Transactions on Cybernetics , 2017 , 47 ( 6 ): 1540 － 1550 .

LeCun Y , Bengio Y , Hinton G . Deep learning [J]. Nature , 2015 , 521 ( 7553 ): 436 － 444 .

Shi C M , Panoutsos G , Luo B , et al . Using multiple-feature-spaces-based deep learning for tool condition monitoring in ultraprecision manufacturing [J]. IEEE Transactions on Industrial Electronics , 2019 , 66 ( 5 ): 3794 － 3803 .

Tsygvintsev A . On the overfly algorithm in deep learning of neural networks [J]. Applied Mathematics and Computation , 2019 , 349 : 348 － 358 .

Alharbi A , Doncker E D . Twitter sentiment analysis with a deep neural network: An enhanced approach using user behavioral information [J]. Cognitive Systems Research , 2019 , 54 : 50 － 61 .

Young T , Hazarika D , Poria S , et al . Recent trends in deep learning based natural language processing [J]. IEEE Computational Intelligence Magazine , 2018 , 13 ( 3 ): 55 － 75 .

舒坚 , 张学佩 , 刘琳岚 , 等 . 基于深度卷积神经网络的多节点间链路预测方法 [J]. 电子学报 , 2018 , 46 ( 12 ): 2970 － 2977 .

Shu J , Zhang X P , Liu L L , et al . Multi-nodes link prediction method based on deep convolution neural networks [J]. Acta Electronica Sinica , 2018 , 46 ( 12 ): 2970 － 2977 . (in Chinese)

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Application of Hadamard ECOC in Multi-Class Problems Based on SVM

A Simplification to Support Vector Machine for Complicated Recognition Problem

Some New Developments on Support Vector Machine

A Method for Continuous Detection and Classification of Malicious Network Traffic Based on Double-Layer Model and Distribution of Indexes

Related Author

YIN An-rong

XIE Xiang

KUANG Jing-ming

FANG Jing-long

CHEN Shuo

PAN Zhi-geng

LIANG Rong-hua

QI Feng-yan

Related Institution

Department of Electronics Engineering,Beijing Institute of Technology

Institute of Graphics and Image,Hangzhou Dianzi University

State Key Laboratory of CAD &amp

Zhejiang University, Hangzhou

College of Information Engineering,Zhejiang University of Technology

⁰